All activity
Charles Sturtleft a comment
AI agents are getting powerful. They can browse the web, execute code, access your files. And we're giving them random plugins from strangers without checking what they do. I spent months scanning 33,530 AI skills across every major marketplace. Found credential stealers, reverse shells, prompt injection attacks — all packaged as "helpful" agent extensions. SkillShield is my answer: a...
SkillShieldThe original AI skills security directory
I built SkillShield after getting paranoid about AI agent security. We're all plugging our agents into random "skills" without knowing what they actually do. So I scanned 33,530 of them.
The results were horrifying: 32% contain critical security issues — credential theft, reverse shells, prompt injection attacks, and data exfiltration
I launched the first version of SkillShield on February 6th after months of scanning and research.
SkillShieldThe original AI skills security directory
Charles Sturtleft a comment
Interesting - I launched skillshield.dev on Feb 6 with the same concept. Would love to chat about how our approaches differ...
SkillShieldSecurity-scored directory for AI skills and agent tools